In a recent National Institutes of Health consensus development conference, it was emphasized that during the 1990's, more than 1.5 million women in the U.S. will be diagnosed with breast cancer and that approximately 30% of them will die of the disease. It is estimated that 1 out of every 9 women will develop breast cancer during her lifetime. In general, the smaller the lesion is at the time of detection, the better the prognosis. Magnetic Resonance Imaging (MRI) offers exciting potential for increased tissue characterization compared to other imaging modalities. The initial goal of this proposal was to improve the characterization of lesions in the breast as either benign or malignant based on an analysis of their metabolism using localized magnetic resonance spectroscopy (MRS). The MRS data collected on more than 50 women show high sensitivity and specificity. All of the women studied had lesions which enhanced with the injection of MR contrast agent. It should be noted that non-enhancing lesions are almost certainly benign. We have also showed that higher fields, i.e. 3 T improve the quality of the MRS examination in two distinctly different ways; improved signal-to-noise in the detection of choline, the metabolite which appears to be a marker for malignancy, and smaller sampling volumes. Over the past year we have developed and validated a multivoxel, multi-slice, MRS sequence for breast MRS at 3T that can provide coverage of the whole breast with about 6 mm in plane resolution in about 10 minutes. This sequence can produce choline maps of the breast that can be readily overlaid on both the anatomical MR images, contrast enhanced MR images or the results of pixel-by-pixel analyses of dynamic contrast studies. Our central hypothesis is that the diagnostic information derived from MRS alone, or in combination with MRI criteria, can be employed to improve the specificity of the overall MR examination. We will test this hypothesis in a cohort of women with enhancing lesions who will all be biopsied as part of this study. Histopathology will be the gold standard for diagnosis. MRS derived improvements in specificity will have a beneficial effect by potentially reducing the number of negative biopsies performed. We will determine the cost-effectiveness of MRS in the overall work-up of breast cancer.